10-04-2013 07:34 AM - edited 03-04-2019 09:13 PM
Hi,
We have 3 routers connecting the office - servers - dr in a triangle, all running eigrp to advertised connected networks...see attached picture
The problem is when the link between the office and the server farm fails traffic should go through dr and reach the servers, but it doesnt. When it happens the DR networks have access to Servers networks, also OFFICE networks have acces to DR networks. I've already tried with floating static route with no luck.
Please see part of the configurations below...just ignore crypto incomplete commands
Any help would be appreciated
SERVERFARM
interface Loopback0
ip address 172.24.31.254 255.255.255.0
!
interface FastEthernet0/0
description SERVERS-OFFICE
ip address 10.50.0.2 255.255.255.252
load-interval 30
speed auto
full-duplex
crypto ipsec df-bit clear
hold-queue 4096 in
hold-queue 4096 out
!
interface FastEthernet0/1
description SERVERS-DR
ip address 10.50.0.10 255.255.255.252
load-interval 30
speed auto
full-duplex
crypto ipsec df-bit clear
hold-queue 4096 in
hold-queue 4096 out
!
router eigrp 200
redistribute connected
redistribute static
network 10.50.0.0 0.0.0.3
network 10.50.0.8 0.0.0.3
network 10.50.0.12 0.0.0.3
no auto-summary
eigrp stub connected static
neighbor 10.50.0.1 FastEthernet0/0
neighbor 10.50.0.9 FastEthernet0/1
!
ip route 0.0.0.0 0.0.0.0 172.24.31.8
OFFICE
interface Loopback0
ip address 172.24.23.254 255.255.248.0
!
interface Loopback1
ip address 172.24.27.254 255.255.252.0
!
interface Loopback2
ip address 172.24.28.254 255.255.255.0
!
interface Loopback3
ip address 172.24.30.254 255.255.255.0
!
interface Loopback4
ip address 10.44.44.254 255.255.255.0
!
interface FastEthernet0/0
description OFFICE-SERVERS
ip address 10.50.0.1 255.255.255.252
load-interval 30
speed auto
full-duplex
crypto ipsec df-bit clear
hold-queue 4096 in
hold-queue 4096 out
!
interface FastEthernet0/1
description OFFICE-DR
ip address 10.50.0.5 255.255.255.252
ip mtu 1400
ip tcp adjust-mss 1300
load-interval 30
speed auto
full-duplex
crypto ipsec df-bit clear
hold-queue 4096 in
hold-queue 4096 out
!
router eigrp 200
redistribute connected
redistribute static
network 10.50.0.0 0.0.0.3
network 10.50.0.4 0.0.0.3
network 10.50.0.12 0.0.0.3
no auto-summary
eigrp stub connected static
neighbor 10.50.0.2 FastEthernet0/0
neighbor 10.50.0.6 FastEthernet0/1
DR
interface Loopback0
ip address 10.221.128.254 255.255.255.0
!
interface FastEthernet0/0
description DR-SERVERS
ip address 10.50.0.9 255.255.255.252
load-interval 30
speed auto
full-duplex
crypto ipsec df-bit clear
hold-queue 4096 in
hold-queue 4096 out
!
interface FastEthernet0/1
description DR-OFFICE
ip address 10.50.0.6 255.255.255.252
ip tcp adjust-mss 1300
load-interval 120
speed auto
full-duplex
crypto ipsec df-bit clear
hold-queue 4096 in
hold-queue 4096 out
!
router eigrp 200
redistribute connected
redistribute static
network 10.50.0.4 0.0.0.3
network 10.50.0.8 0.0.0.3
distribute-list EIGRP_OUT out
no auto-summary
eigrp stub connected static
neighbor 10.50.0.10 FastEthernet0/0
neighbor 10.50.0.5 FastEthernet0/1
Thanks,
10-04-2013 07:40 AM
Can you post the route-map EIGRP_OUT from the DR router? If you look at the eigrp topology table on all three routers, does each router have two paths for each subnet?
HTH,
John
*** Please rate all useful posts ***
10-04-2013 07:51 AM
ip access-list standard EIGRP_OUT
permit 192.168.1.32 0.0.0.3
No ... only 1 route
10-04-2013 09:43 AM
Hello
Why redistribute connected and static and all 3 sites eigrp stubs?
If redistribute is the wqy you want to go with stub areas try a route map with default originate and allowed redistributed routes also
res
paul
Sent from Cisco Technical Support Android App
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide